What problem does it solve? Engineering skills like to-issues, triage, diagnose, and tdd need to know where your repo tracks issues, which triage label strings you use, and where your domain documentation lives. Without this per-repo configuration, those skills guess wrong about your workflow and produce mismatched output. ## Core Features & Use Cases - Guided repo exploration: Inspects git remotes, existing AGENTS.md/CLAUDE.md, CONTEXT.md, ADR directories, and .scratch/ to understand the current state before writing anything. - Three-way configuration: Walks you through choosing an issue tracker (GitHub via gh, GitLab via glab, local markdown under .scratch/, or a custom workflow), mapping the five canonical triage labels, and declaring single-context vs multi-context domain docs. - Scaffolded output: Writes an ## Agent skills block into your existing CLAUDE.md or AGENTS.md plus seed files under docs/agents/ (issue-tracker.md, triage-labels.md, domain.md). - Use Case: You just cloned the Matt Pocock engineering skills into a GitLab-hosted monorepo. Run this skill once so triage applies your real label names and improve-codebase-architecture reads the correct per-context CONTEXT.md files. ## Quick Start Ask the agent to run the setup-matt-pocock-skills skill to configure the issue tracker, triage labels, and domain docs for this repository.
